Request for review of gaps document

[dan@xxxxxxxxxxx (Dan Kohn)]

Some additional comments.

           - dan
-- 
Dan Kohn <mailto:dan@xxxxxxxxxxx>
COO, The Linux Foundation <http://www.linux-foundation.org>
<http://www.dankohn.com/>  <tel:+1-415-233-1000>

---------- Forwarded message ----------
From: James Bottomley <James.Bottomley@xxxxxxxxxxxxxxxxxxxxx>
Date: Nov 15, 2007 9:18 AM


On Thu, 2007-11-15 at 09:59 -0700, Jonathan Corbet wrote:
> Some of the rest (block
> checksums, fast/writable snapshots) is called Btrfs, which would
> certainly come along faster if they put some resources into helping
> its
> development.

Actually, this is something called block guard ... they want the ability
to checksum through the I/O path down to the disk.  Oracle wants the
same functionality, and is working on it (hopefully btrfs will use it
when it's in kernel).

> Persistent shared memory: something along these lines could be
> implemented now with SYSV SHM and a process to save it to persistent
> storage and bring it back.  If they want a region of RAM which remains
> untouched over unexpected reboots and is waiting for them when the
> system comes back, that's not there now.  It could be implemented
> without *too* much trouble, I think - it would be a mechanism not to
> dissimilar to the crash dump kernel stuff that's done now.  Merging it
> could require some persuasion.

I'm wondering if this isn't perhaps the ability to mmap the nvram that a
lot of systems now come with?  If it is, it should be trivial, it's just
that the various nvram controllers require drivers (Intel's HECI one has
just gone over the kernel list).


> Layer 2 tunneling: I can't really speak to this one.  Ask Mr.
> Hemminger,
> he'll know.

There's a llinux OpenL2TP project on SF.net:

http://openl2tp.sf.net

The actual driver, which is a PPP over L2TP is already in the kernel ...
is there a more generic protocol they need?

James